KR20160006703A - Method, apparatus and system for isolating microphone audio - Google Patents

Abstract

A method, apparatus and system for isolating microphone audio includes recording audio using at least two microphones using a target microphone of an array of microphones, using the target microphones of the array of microphones, Determining an attenuation factor for the resulting audio, determining a delay factor for audio resulting from individual positions of other microphones of the array of microphones, and capturing by the target microphone to separate the audio signal captured by the target microphone And implementing a determined attenuation factor and a delay factor to remove audio from the individual positions of other microphones from the audio signal. The method, apparatus, and system may include processing a separate audio signal of a target microphone to determine audio attributes of a separate audio signal of a target microphone, and processing individual sources of audio in the separate audio signal using audio attributes And < / RTI >

Description

[0001] METHOD, APPARATUS AND SYSTEM FOR ISOLATING MICROPHONE AUDIO [0002]

This application is related to International PCT Application No. PCT / US12 / 072083, filed December 28, 2012, the entire contents of which are incorporated herein by reference for all purposes.

The present invention relates generally to the separation of microphone audio, and more particularly, to a method, apparatus, and system for removing noise from microphone signals to separate audio.

Noise suppression is often required to suppress noise in multiple communication systems and content distribution devices, improving communication quality and media understanding. Noise suppression may be achieved using a variety of techniques, some of which may be classified as single microphone technologies and array microphone technologies.

Array microphone noise reduction techniques use multiple microphones placed at different locations to form a beam and separated by a certain minimum distance from each other. Typically, the beam is used to capture speech, which is then used to reduce the amount of noise captured outside the beam. Thus, array microphone technologies can suppress non-stationary noise. Separation of the microphone signals through noise suppression may be used in a retail advertising environment, for example, to identify shopper demographics and / or purchase numbers.

However, multiple microphones also generate more noise per se. Additionally, such techniques do not use the system's configuration parameters and known audio signals to enable noise cancellation as described herein.

Embodiments of the present invention address the drawbacks of the prior art by providing a method, apparatus, and system for isolating microphone signals.

In one embodiment of the present invention, a method includes recording audio using at least two microphones using a target microphone of an array of microphones, using the target microphones of the array of microphones, Determining an attenuation factor for the audio that has occurred, determining a delay factor for the audio resulting from the individual positions of other microphones of the array of microphones, and capturing by the target microphone to separate the audio signal captured by the target microphone And implementing a determined attenuation factor and a delay factor to remove audio from the individual positions of other microphones from the audio signal. The method, apparatus, and system may include processing a separate audio signal of a target microphone to determine audio attributes of a separate audio signal of a target microphone, and processing individual sources of audio in the separate audio signal using audio attributes And < / RTI >

In an alternative embodiment of the present invention, an apparatus includes a memory for storing program routines and data, and a processor for executing program routines. In such an embodiment, the apparatus records audio using at least two microphones comprising an array of microphones and uses the target microphones of the array of microphones to generate audio from the individual positions of the other microphones of the array of microphones Determine a delay factor for audio generated from individual positions of other microphones of the array of microphones using the target microphone of the array of microphones, and determine a delay factor for the audio signal captured by the target microphone Implements the determined attenuation and delay factors to remove audio from the individual positions of other microphones of the array of microphones from the audio signal captured by the target microphone Ridoen and process the audio signal separated in the target microphone in order to determine the audio properties of the audio signal, using the audio attribute, and is configured to determine the individual audio source of the audio signal of a target in a separate microphone.

In an alternative embodiment of the present invention, a system includes an apparatus comprising at least two microphones comprising an array of microphones, at least one audio source, program routines and a memory for storing data, Lt; / RTI > In such a system, the apparatus records audio using at least two microphones, including an array of microphones, and uses the target microphones of the array of microphones to generate audio data for audio generated from individual positions of other microphones of the array of microphones Determine an attenuation factor, determine a delay factor for audio generated from individual positions of other microphones of the array of microphones using the target microphone of the array of microphones, and determine a delay factor for the audio signals captured by the target microphone Implement a determined attenuation factor and a delay factor to remove audio from the individual positions of other microphones of the array of microphones from the audio signal captured by the microphone, The process the audio signal separated in the target microphone in order to determine the audio properties of the audio signal, using the audio attribute, and is configured to determine the individual audio source of the audio signal of a target in a separate microphone.

The teachings of the present invention can be readily understood by considering the following detailed description together with the accompanying drawings.
1 shows a high level block diagram of a content distribution system to which an embodiment of the present invention may be applied.
2 shows a high-level block diagram of an in-store ad network for providing in-store advertising to which embodiments of the present invention may be applied.
Figure 3 shows a high level block diagram of an apparatus for separating microphone audio, in accordance with an embodiment of the present invention.
Figure 4 shows a flow diagram of a method for separating microphone audio, in accordance with an embodiment of the present invention.
It is to be understood that the drawings are for purposes of illustrating the concepts of the invention and are not necessarily the only possible configurations for illustrating the invention. In order to facilitate understanding, the same reference numerals have been used, where possible, to designate the same elements that are common to the figures.

The present invention advantageously provides a method, apparatus and system for isolating microphone audio. Although the present invention will be described primarily in the context of in-store retail advertising network environment and advertising content distribution and specifically checkout applications for separating speech, certain embodiments of the present invention are intended to be illustrative only, . It will be appreciated by those skilled in the art that the concepts of the present invention may be advantageously applied to any environment where separation of any audio, such as voices, such as fast food restaurants, bank teller counters, .

The functions of the various elements shown in the figures may be provided through use of dedicated hardware as well as hardware capable of executing software in association with appropriate software. When provided by a processor, the functions may be provided by a single dedicated processor, by a single shared processor, or by a plurality of individual processors, some of which may be shared. Furthermore, the explicit use of the term " processor "or" controller "should not be construed to refer exclusively to executable hardware, ROM "), random access memory (" RAM "), and non-volatile storage. Moreover, all statements herein reciting principles, aspects, and embodiments of the invention, as well as specific examples thereof, are intended to cover both structural and functional equivalents thereof.

Additionally, such equivalents are intended to include both currently known equivalents as well as equivalents developed in the future (i. E., Any developed elements that perform the same function regardless of structure).

Thus, for example, it will be appreciated by those skilled in the art that the block diagrams presented herein represent conceptual views of exemplary system components and / or circuits embodying the principles of the invention. Similarly, any flow charts, flow diagrams, state transition diagrams, pseudo code, etc., may be stored on a computer readable medium, such as a computer readable medium, It will be appreciated that the various processes that may be implemented are represented.

1 shows a high level block diagram of a content distribution system to which an embodiment of the present invention may be applied. The content distribution system 100 of FIG. 1 illustratively includes a single server 110, a plurality of receiving devices, such as tuning / decoding means (illustratively, set top boxes (STBs) 120 ₁ -120 _n And a separate display 130 _{1 -} 130 _n for each of the set-top boxes 120 _{1 -} 120 _n , respectively. 1, display 130 includes separate microphones 132 ₁ -132 _n and at least one speaker 133 ₁ -133 _n , respectively, and individual checkout lanes 134 ₁ -134 _n , . In the content distribution system 100 of FIG. 1, the microphones 132 of the displays 130 include an array of microphones. In such a system, such as the system 100 of Figure 1, the microphones 132 are typically used to confirm playout of the content onto the displays 130 and may further be used for noise cancellation purposes.

Although each of the plurality of set-top boxes 120 _{1 -} 120 _n in the system 100 of FIG. 1 is illustratively connected to a single discrete display, in alternate embodiments of the present invention, Each of the display devices 120 _{1 -} 120 _n may be connected to more than one display. That is, in alternative embodiments of the present invention, displays of a plurality of checkout lanes may be controlled and communicated to a single set-top box. Additionally, although tuning / decoding means is illustratively illustrated as set-top boxes 120 in the content distribution system 100 of FIG. 1, in alternative embodiments of the present invention, the tuning / decoding means of the present invention Decoding means such as tuning / decoding circuitry integrated into displays 130 or other standalone tuning / decoding devices, and the like. Still further, the receiving devices of the present invention may include any device capable of receiving content such as audio, video, and / or audio / video content.

In one embodiment of the present invention, the content distribution system 100 of Figure 1 may be part of an in-store advertising network. For example, FIG. 2 shows a high-level block diagram of an in-store ad network 200 for providing in-store advertising. 2, the ad network 200 and the distribution system 100 may be used to store music recordings, home videos, product (s), < RTI ID = 0.0 & Descriptions, advertising content, and other such content in the form of a combination of hardware and software that provides cataloging, distribution, presentation, and usage tracking. The content may include content presented in a compressed or uncompressed video and audio stream format (e.g., MPEG4 / MPEG4 Part 10 / AVC-H.264, VC-1, Windows Media, etc.) But should not be limited to using only those formats.

In one embodiment of the invention, the software for controlling the various elements of in-store advertising network 200 and content distribution system 100 is a 32-bit operating system (e.g., MS Windows < (R) > X Windows operating system) and high performance computing hardware. The ad network 200 may utilize a distributed architecture and, in one embodiment, may include a satellite (or other method, e.g., a wide area network (WAN), the Internet, a series of microwave links, or similar mechanisms) It provides centralized content management and distribution control through my modules.

2, the content for the in-store ad network 200 and the content distribution system 100 may include advertiser 202, recording company 204, movie studio 206, or other content providers 208, / RTI > The advertiser 202 may be an advertisement company, or other entity that represents a product manufacturer, service provider, manufacturer, or service provider. The ad content from advertiser 202 may be made up of audiovisual content including commercial, "info-mercial ", product information, product description, and the like.

The recording company 204 may be a record label, a music publisher, a licensing / publishing entity (e.g., BMI or ASCAP), a personal artist, or other such source of music related content. Recording company 204 provides audiovisual content such as music clips (short segments of recorded music), music video clips, and the like. The movie studio 206 may be a movie studio, a film production company, a public relations person, or other source related to the movie industry. The movie studio 106 may provide movie clips, pre-recorded interviews with unedited actors, movie reviews, behind-the-scenes presentations, and similar content.

Other content providers 208 may be any other provider of video, audio, and audiovisual content that may be distributed and displayed, for example, through the content distribution system 100 of FIG.

In one embodiment of the present invention, the content is obtained via the network management center 210 (NMC) using, for example, traditional recorded media (tapes, CDs, videos, etc.). The content provided to the NMC 210 is edited, for example, in a form suitable for distribution to a local distribution system 100 that distributes and displays the content to a local site.

The NMC 210 may digitize the received content and provide the content to the network operations center (NOC) 220 in the form of digitized data files 222. It will be noted that although the data files 222 are mentioned in terms of digitized content, they may also be streaming audio, streaming video, or other such information. The content edited and received by the NMC 210 may include commercials, bumpers, graphics, audio, and the like. All files are preferably named to be uniquely identifiable. More specifically, the NMC 210 creates distribution packs targeted to specific sites, such as store locations, and delivered to one or more stores on a scheduling or order basis. The distribution packs, if used, include content intended to replace or enhance existing content that already exists locally (if the site's system is not being initialized for the first time, then the delivered packages will be the initial Will form the basis of the content). Alternatively, the files may be compressed and transmitted separately, or some type of streaming compression program is employed.

The NOC 220 communicates the digitized data files 222 to the content distribution system 100 at the commercial sales outlet 230, in one example, over the communications network 225. The communication network 225 may be implemented in any one of several technologies. For example, in one embodiment of the present invention, a satellite link may be used to distribute the digitized data files 222 to the content distribution system 100 of the commercial sales outlet 230. This allows the content to be easily distributed by broadcasting (or multicasting) the content to various locations. Alternatively, the Internet may be used to distribute audiovisual content to the commercial outlet 230 and to allow feedback from the commercial outlet 230. Other ways of implementing the communication network 225, such as using leased lines, microwave networks or other such mechanisms, may also be used in accordance with alternative embodiments of the present invention.

The server 110 of the content distribution system 100 is capable of receiving content (e. G., Distribution packs) and thereby storing them in a variety of receivers, such as set-top boxes 120 and displays 130 Lt; / RTI > That is, in the content distribution system 100, the content is received and configured for streaming. Streaming may be performed by one or more servers configured to operate together or together. The streaming content may include content configured for a variety of different locations or products across a sales outlet 230 (e.g., a store). For example, the individual set-top boxes 120 and displays 130 may be located at specific locations throughout the sales outlet 230 and may be located within a predetermined distance from the location of each individual set- And to broadcast the audio, respectively.

Various embodiments of the present invention provide a method, apparatus, and system for isolating microphone signals. That is, the various embodiments of the invention described herein are directed to removing ambient noise from a signal of a microphone present in a commercial checkout environment such that the audio or sounds originating from the individual checkout counter can be separated. More specifically, the various embodiments of the present invention described herein may be applied to a plurality of display screens, e. G., As shown in Fig. ≪ RTI ID = 0.0 > 1, < To remove ambient sounds from the microphones contained in the microphone. Also, although the various embodiments of the present invention will be described primarily in the context of a commercial ad network environment and ad content distribution, certain embodiments of the invention should not be viewed as limiting the scope of the invention.

In one embodiment of the present invention, a method for determining noise, such as sounds and other audio signals, generated in adjacent checkout lanes of the content distribution system of Fig. 1 to be removed from at least one microphone in an array of microphones The process may be accomplished through a beam forming process / technique, in an embodiment of the present invention. To illustrate one embodiment of the invention, t is the time slot in which the microphones record sound (e.g., every msec), and y _i (t) is the time slot at time (t) X _i (t) is a sound signal generated at the counter i in the time slot t (e.g., between the cash teller at the counter (i) and the customer, T _ij is a weighted value (delay parameter) based on a time delay from the counter (i) to the counter (j), and w _ij is the counter (i) j) based on the distance between them (attenuation factor). Accordingly, the microphone at position (i) receives a signal y _i containing sounds from all counters that can be determined according to equation (1) according to:

Further, in the equation (1), w _ji is the attenuation factor from the counter j to the counter i and T _ij is the delay parameter from the counter j to the counter i. As a result, in order to separate the sound coming from the counter i, the following processing occurs. Each of the displays may be configured to provide the recorded signals y _i (t), for example, in various embodiments of the present invention, to the set-top box 120, or to the server of the content distribution system 100 110 or a processing device that may reside in a local or remote server, such as the NMC 210 or the NOC 220 of the in-store advertising network 200 of FIG. Having these signals, the processing device solves the linear system of equation (1) to separate the sound (i. E., X _i (t)) at the counter _i at time t. In this system, the unknowns are the signals (x _i) in different time slots (t).

3 is a block diagram of an exemplary embodiment of a set-top box 120 or a server 110 of the content distribution system 100 of FIG. 1 or an NMC 210 of the in-store advertising network 200 of FIG. 2, in various embodiments of the present invention. Level block diagram of a processing device, which may be a local or remote server, such as the NOC 220, or the NOC 220. More specifically, the processing device of FIG. 3 illustratively includes a memory 320 for storing control programs, file information, stored signals, etc., as well as the processor 310. The processor 310 cooperates with conventional support circuitry 330, such as power supplies, clock circuits, cache memory, etc., as well as circuits that help to execute software routines stored in the memory 320. [ Accordingly, it is contemplated that some of the process steps discussed herein as software processes may be implemented in hardware, for example, as circuits that perform various steps in cooperation with the processor 310. [ The processing device also includes an input-output circuit 340 that forms an interface between the various functional elements in communication with the processing device.

Although the processing apparatus of FIG. 3 is illustrated as a general purpose computer programmed to perform various control functions according to the present invention, the present invention may be implemented in hardware, for example, as an application specific integrated circuit (ASIC). Accordingly, the process steps described herein are intended to be broadly interpreted as being performed evenly by software, hardware, or a combination thereof, executed by a processor. Additionally, although the processing device of FIG. 3 is shown as a separate component, the functions of the processing device according to the inventive concepts and embodiments described herein may be incorporated into existing system components such as a set top box, server, .

Returning to equation (1) above, in one embodiment of the present invention, to determine the attenuation factor w _ij and the delay factor T _ij , Known checkout sounds or tones generated by the user are used. That is, in such an embodiment, the check-out scanner tone is a known sound and comprises a predetermined volume. When each scanner generates a checkout tone at a known time t ₁ , the microphone of the target display detects the tones and transmits such information to the processing of the present invention, for example, Lt; RTI ID = 0.0 > device / server. ≪ / RTI >

In an alternative embodiment of the present invention in which local sounds are not known (i.e., the type and volume of locally generated audio is not known), a local (local) sound, such as microphone 132 ₁ of the individual checkout lane 134 ₁ , The microphone can be used to record audio signals in the vicinity thereof and using known techniques such as beam forming techniques and other audio signal processing techniques and can determine which audio signals are locally generated nearby, The volume and other physical characteristics of such locally generated audio signals. These determined parameters of the locally generated audio signals can then be used by the target microphone to determine the attenuation and delay factors of such signals as described above. That is, in such embodiments, the locally generated audio signals, as determined by the individual microphones of the array, may be combined with known signals such as those described above to determine the attenuation and delay factors of such signals as described above Lt; RTI ID = 0.0 > microphone.

In one embodiment of the invention, the audio circuitry may include, for example, a separate circuit card in the display or server of the present invention, or may include a circuit card as described in co-pending U.S. Patent Application No. 12 / 733,214, And may include a dedicated device such as an audio processor. The audio circuit of the present invention can calculate an attenuation factor (w _ij ) and a delay factor (T _ij ) for each scanner at each checkout counter having information about known sounds generated at checkouts have.

More specifically, in an embodiment of the present invention, considering that a scanning signal at position i is generated at time t ₁ , T _ij is the time slot in which the scanning signal is recorded first in microphone j lt; _{RTI ID = 0.0} > t1. < / RTI > Alternatively, in an alternative embodiment of the present invention, the difference in time slots between the first / highest peaks over different recorded signals may be used rather than the beginning of the signal.

In one embodiment of the present invention, the attenuation factor w _ij is similarly calculated. In particular, w _ii may be taken to be equal to 1 for all i. The factor w _ij is calculated as the ratio of the signal at the microphone j at time t ₁ + T _ij to the signal at the microphone i at time t ₁ + T _ii . In an alternative embodiment of the invention, the peaks in the waveform of the scanning sound or the proportion of other positions may be used.

Once the attenuation factor w _ij and the delay factor T _ij are computed, a beamforming technique may be used so that sounds from other checkout counters are received by the target microphone in the target display 100 Lt; / RTI >

In various embodiments of the present invention, once the ambient noise has been removed from the received audio signal, for example, in the target display 110, as described above, multiple processes separate the desired audio, such as speech Lt; / RTI > For example, detection and separation of the speech of the customer and the money teller near the target display 110 may be required. In such a case, the teller is typically assumed to speak first after a series of audio tones representing items purchased. The money teller is also assumed to make repetitive statements such as "Your total amount is ...", "You saved ...", "Your wife", "Teacher"

In one embodiment of the present invention, the following audio attributes can be detected or determined by performing a Fourier transform on audio signals, such as audio representing a dialog between a teller and a customer:

a. Frequency

b. Average amplitudes

c. Maximum amplitudes

d. The time of the first amplitude peak

e. Number of amplitude peaks

f. Assignment of a zero or one indicator of whether a voice signal, snippet, or segment is likely to be a teller or a customer.

In various embodiments of the present invention, such processing may be performed by, for example, the target display 110 and / or the audio card at the central server 140. In various embodiments of the present invention, standard machine learning techniques, such as but not limited to k-means clustering, use the determined audio attributes with at least audio samples to determine which audio samples, for example, And determine which audio samples represent the customer's speech. As described above and according to the described embodiments of the present invention, audio samples, segments or signals generated near the target display 110 may be determined / separated.

Once the audio, such as speech generated by a given customer, is separated, standard machine learning techniques, such as, but not limited to, linear regression, decision trees, AdaBoost (TM) and support vector machines or algorithms, , Information about the audio, for example in the case of speech, can be attempted to determine the gender, age, ethnic background, etc. of the customer. For example, in an embodiment of the invention, a database of training data sets is generated using people of known gender, age and ethnicity based on each person's detected frequencies, amplitudes, frequency magnitude peaks . Thereafter, the training data sets may be used to train functions, algorithms, and / or software modules so that the functions can predict gender, age or ethnic background. It should be noted that it would be advantageous to have people in the control group say certain phrases that are often spoken at checkout counters to help improve the detection of gender, age or ethnicity. It should also be noted that the same process can be applied to audio other than speech, e.g., audible tones associated with the scanning of the product. Moreover, if the actual audio from a particular store for which the method of the present invention should be implemented can be collected and used to generate training data sets, the accuracy of the function can be further improved based on remaining ambient noise, geographic dialect / grammar, This should also be noted.

In alternative embodiments of the present invention, speech-to-text software can be used to detect certain words or phrases such as mom, dad, teacher, miss, etc. that help improve the identification of age, gender or ethnicity have. Additionally, in further alternative embodiments of the present invention, separation of eye crying, gnashing, etc. may be used to assume the existence of a family. Determination of other purchasing information, such as purchasing information, such as customer attributes, including age, gender, ethnicity, family, etc., and audible tones associated with the scanning of products, in accordance with various embodiments of the present invention described herein, May be used to provide advertisements and ad (s) to the customer (s) via the target display 110, for example.

In alternative embodiments of the present invention, the audio / speech information determined from the display microphones as described above is combined with data (e.g., scanned items, customer card information, etc.) collected by the retail environment Thereby increasing the accuracy of identifying customers' gender, age, and / or other demographic information. In various embodiments of the present invention, combining the determined customer information with, for example, time stamp information can yield highly valuable information. For example, if a woman is found shopping at certain times of the day, the ad may be shifted to deliver more appropriate ads to the woman during those times.

In one embodiment of the present invention, once a clean audio pattern of speech is determined, the audio pattern is used to produce a voice print. The gates can then be used to identify the shopper. For example, a significant value is obtained by observing the pattern of visit to a store. If a given gateways can be traced to establish shopper patterns such as the fact that shoppers visit every Tuesday, once a week, or every other Wednesday, the data is of high value. Collecting data from all detected gates can be used to establish overall patterns of shopper frequency. This data can then be used to optimize the advertising period and refresh dates. For example, if the data indicates that shoppers typically visit twice a week and the media is required to be fresh on each visit, the rate at which the new media is refreshed may be increased.

According to various embodiments described herein, once a shopper is identified by a gated statement as described above, the shopper can always be identified using that gated statement, even if the shopper is only pseudo-identified. In alternative embodiments of the present invention, for example, shopper information collected by, for example, a store using a customer card may be used to further identify the shopper.

In alternative embodiments of the present invention, in addition to the speech as described above, the audio in the separate audio signal of the target microphone can be used to provide targeted advertising and ads, for example, In order to improve the effectiveness of the advertisement by obtaining information about the purchase transaction. More specifically, in one embodiment of the present invention, the audio tones associated with the scanning of the item to be purchased can be recorded by the microphone of the target display and used to determine a plurality of items purchased by a particular customer. Additionally, such information may be combined with information held by the retailer, for example about what items were purchased at a particular time in a particular register, and the specific items purchased may be associated with a particular customer.

In various embodiments of the present invention, the separated audio recorded by the microphone, which has been separated as described above, can be used to display targeted advertisements and advertisements, for example, as described above, (S) to obtain information about the purchase transaction to improve the effectiveness of the advertisement.

Figure 4 shows a flow diagram of a method for separation of microphone audio, in accordance with an embodiment of the present invention. The method 400 of FIG. 4 begins at step 402 where environmental sounds / audio are recorded by at least two microphones, including an array of microphones. The method 400 proceeds to step 404.

At step 404, an attenuation factor for sounds from all other microphones of the array other than the microphone being calibrated (i.e., the target microphone) is calculated using known sounds from, for example, the positions of other microphones of the array . The method 400 proceeds to step 406.

At step 406, a delay factor for sounds from all other microphones of the array other than the microphone to be calibrated (i.e., the target microphone) is calculated using, for example, known sounds from the positions of other microphones of the array . The method 400 proceeds to step 408.

In step 408, the determined attenuation factor and delay factor may be determined, for example, by using the beam forming processes / techniques in an embodiment of the present invention, to separate the audio signal captured by the target microphone And to remove audio from separate positions of other microphones of the array of microphones from the audio signal from the audio signal captured by the target microphone. The method 400 proceeds to step 410.

In step 410, the separated audio signal of the target microphone is processed to determine the audio attributes of the separated audio signal of the target microphone. For example and as described above, in one embodiment of the present invention, speech, such as frequency, average amplitude, maximum amplitude, time of the first amplitude peak, and amplitude peaks in the isolated speech of the target microphone May be determined by performing a Fourier transform on the separated audio signals. The method 400 then proceeds to step 412.

In step 412, individual sources of audio in the separated audio signal of the target microphone are determined using audio attributes. As described above, in one embodiment of the present invention, the sources of speech in the separate audio signal of the target microphone are determined by applying standard machine learning techniques to the separated audio signal and applying the determined speech attributes . The method 400 may then proceed to optional steps 414 or 416 or may terminate.

In optional step 414, a standard machine learning technique may be used to determine demographic information, such as gender, age, ethnic background, etc., of at least one individual source of speech, at least one of the separate sources of audio, Signal.

In optional step 416, the targeted advertisement is directed to at least one of the determined individual sources of audio. For example, as described above, in one embodiment of the present invention, targeted advertisements and advertisements may be presented to the identified / determined customer (s), e.g., via a target display.

While various embodiments of a method, apparatus, and system for isolating microphone audio (intended to be exemplary and not limiting) have been described, it is noted that variations and variations may be practiced by those skilled in the art in light of the above teachings do. It is therefore to be understood that changes may be made in the specific embodiments of the invention disclosed which are within the scope and spirit of the invention. While the foregoing is directed to various embodiments of the present invention, other and further embodiments of the invention may be devised without departing from the basic scope thereof.

Claims (20)

Recording audio using at least two microphones comprising an array of microphones;
Using the target microphone of the array of microphones to determine an attenuation factor for audio generated from individual positions of other microphones of the array of microphones;
Using a target microphone of the array of microphones to determine a delay factor for audio resulting from individual positions of other microphones of the array of microphones;
The attenuation factor and the delay factor determined to remove audio from the individual positions of the other microphones of the array of microphones from the audio signal captured by the target microphone to isolate the audio signal captured by the target microphone, ;
Processing the separated audio signal of the target microphone to determine audio attributes of the separated audio signal of the target microphone; And
And using the audio properties to determine individual sources of audio in the separated audio signal of the target microphone. The method according to claim 1,
Wherein the audio attributes include speech attributes and individual sources of speech in the separated audio signal of the target microphone are determined. 3. The method of claim 2,
Wherein the processing comprises applying a Fourier transform to the separated audio signal of the target microphone to determine attributes of speech in the audio signal. The method of claim 3,
Wherein the attributes of the speech include at least one of a frequency, an average amplitude, a maximum amplitude, a time of a first amplitude peak, and a number of amplitude peaks. 3. The method of claim 2,
Determining individual sources of speech in the separated audio signal comprises applying a machine learning technique to the separated audio signal and applying the determined speech attributes. 6. The method of claim 5,
Wherein the machine learning technique comprises k-means clustering. 3. The method of claim 2,
Applying a standard machine learning technique to the separated audio signal of at least one of the individual sources of the speech to determine demographic information of at least one respective source of speech. 8. The method of claim 7,
Wherein the standard machine learning technique comprises at least one of linear regression, decision trees, AdaBoost, and support vector machines or algorithms. 8. The method of claim 7,
Wherein the demographic information comprises at least one of a gender, an age, and an ethnic background of the source of the speech. 3. The method of claim 2,
And using the speech attributes to determine a gender for individual sources of the speech. The method according to claim 1,
Wherein the audio characteristics include audio characteristics of the audible tones associated with the purchase of the product and wherein a plurality of purchased products are determined from the audible tones. The method according to claim 1,
Identifying individual sources of the audio in the separated audio signal of the target microphone using information collected by the retailer. The method according to claim 1,
And providing targeted advertisements for determined individual sources of audio. As an apparatus,
A memory for storing program routines and data; And
A processor for executing the program routines;
The apparatus comprises:
Recording audio using at least two microphones including an array of microphones;
Determining, using a target microphone of the array of microphones, an attenuation factor for audio generated from individual positions of other microphones of the array of microphones;
Using a target microphone of the array of microphones to determine a delay factor for audio generated from individual positions of other microphones of the array of microphones;
The attenuation factor and the delay factor determined to remove audio from the individual positions of the other microphones of the array of microphones from the audio signal captured by the target microphone to isolate the audio signal captured by the target microphone, ≪ / RTI >
Processing the separated audio signal of the target microphone to determine audio attributes of the separated audio signal of the target microphone; And
Using the audio properties to determine individual sources of audio in the separated audio signal of the target microphone
Lt; / RTI > 15. The method of claim 14,
Wherein the device comprises at least one integrated audio circuit of a server and a set-top box. As a system,
At least two microphones comprising an array of microphones;
At least one audio source; And
A memory for storing program routines and data, and a processor for executing the program routines,
The apparatus comprises:
Recording audio using at least two microphones including an array of microphones;
Determining, using a target microphone of the array of microphones, an attenuation factor for audio generated from individual positions of other microphones of the array of microphones;
Using a target microphone of the array of microphones to determine a delay factor for audio generated from individual positions of other microphones of the array of microphones;
The attenuation factor and the delay factor determined to remove audio from the individual positions of the other microphones of the array of microphones from the audio signal captured by the target microphone to isolate the audio signal captured by the target microphone, ≪ / RTI >
Processing the separated audio signal of the target microphone to determine audio attributes of the separated audio signal of the target microphone; And
Using the audio properties to determine individual audio sources in the separated audio signal of the target microphone
≪ / RTI > 17. The method of claim 16,
Wherein the at least two microphones comprise at least one network audio processor microphones. 17. The method of claim 16,
Wherein the at least two microphones include microphones in a checkout lane of a retail environment. 17. The method of claim 16,
Wherein the at least one audio source comprises a scanner. 17. The method of claim 16,
Wherein the at least one audio source comprises a cash teller and a customer.

Images